methods of measurement of muscle and joint function · methods of measurement of muscle and joint...
TRANSCRIPT
CRITICAL REVIEW
METHODS OF MEASUREMENT OF MUSCLE AND JOINT FUNCTION
NANCY SALTER, OXFORD, ENGLAND
FrIll?! 1/u’ .‘Ilcdical Research Comnzcil Unit for Research on Climate and 11’ovking Lftmcienci’,
I)e/’arlnzent of .‘�1 )matoniv, U)mim’l’VSitV of Oxford
‘ ‘ Iii the movements of an extremity, the two most important factors involved are the
1)�\t’er or amount of force present and the amplitude or range of motion ; the former beingindicative of the amount of function present in the muscles crossing the joint and the latter
of the amount of motion between the joint surfaces “ (Robinson 1921). It is perhaps necessary
to) add a third equally important factor, the coordination of muscular contraction required
to bring about the desired movement, for without this the limbs are unable to serve a useful
functioiu. This review is specifically concerned with the two criteria of joint function mentioned
by Robinson, although accurate measurements of muscle strength and amplitude of movement
with the voluntary participation of the subject demand normal coordination of the muscle
activity,THE NEED FOR ACCURATE METHODS
Study in this field of medicine received its main impetus from the large number of
orthopaedic cases resulting from the 1914-18 war (Nutter 1919, Albee and Gillilaiid 1920,
(‘lark 1920, Rosen 1922, Silver 1923) and from the poliomvelitis epidemics in the United
States at the beginning of the century (Lovett 1916). Investigations carried out at this time
indicated the iueed for accurate assessment of muscle and joint function and led to the
development of numerous instruments and methods for measuring and recording the required
values. Work ��‘as carried on spasmodically until the second world war, when interest in
such measurements was renewed because of the increased incidence of limb injuries. Also,
�vith the l)re\’�Ilence of J)Ohiomyehitis during the last ten years, further studies have been
umudertaken in connection with the problems of rehabilitation after this disease.
The practical application of the accurate knowledge of the force and amplitude of
movements is apparent in many branches of medicine (Martin and Lovett 1915, Albee and
(;ilhilaiud 1920, Rosen 1922, W’iechec and Krusen 1939, Molander and Weinmann 1942,
Schwab, Watkins and Brazier 1943, Taylor and Brozek 1944, McIntosh, Badgley, Ghormley,
(;tidakuiust, Iv\’, Karsner, Lee and Viets 1945, Duvall 1948, Hellebrandt, Skowlund and
Kelso 1948, and McBride 1950). In clinical work, for example, accurate measurements are
important iii diagnosis and prognosis and as a guide to treatment and its effects. They also
serve to record the ultimate functional recovery or the residual disability and are particularly
useful in deciding when treatment should cease. Repeated measurements during treatment
were recommended by Albee and Gilhiland (1920) in order to arouse and maintain the interest
of the patient in the progress of his own case ; under the term ‘ ‘ metrotherapy ‘ ‘ they included it
as a definite part of treatment. Wiechec and Krusen (1939), considering the amplitude of move-
ments, also supported this view. We have found that patients show great interest in frequent
accurate measurements of the strength of a muscle group or the amplitude of a movement.
In industrial accident cases, accurate records may be required of the extent of the injury
and the degree of functional disability. The just award of compensation may depend on the
accurate assessment of the degree of disability (Rosen 1922).
Accurate measurements omi normal subjects are of fundamental imiterest in the study of
joinut mechanics (Darcus 1951, Salter and Darcus 1952, Darcus and Salter 1953, Salter and
I)arcus 1953), and may he concerned in specific problems which have their application in
clinical work (I)arcus 1953). Accurate data are also of value outside the clinical field; normal
474 THE JOURNAL OF BONE ANI) JOINT SURGERY
METHODS OF MEASUREMENT OF MUSCLE AND JOINT FUNCTION 475
values for the strength and amplitude of movements have become of increasing importance
in the design of equipment and machinery. Certain tests of muscle strength have also been
incorporated in series of ‘ ‘ physical fitness ‘ ‘ tests used by the Services. Yet for whatever
1)tirl)�5e measurements are made, they can only be as useful as they are reliable.
ADVANTAGES AND DISADVANTAGES OF SUBJECTIVE AND OBJECTIVE METHODS
The earliest methods of measuring muscle strength and the amplitude of joint movememuts
were subjective-that is, they were dependent on the personal impressions of an observer.
The need for reliable and more accurate methods has led, in many cases, to the replacement
of subjective estimation by more objective methods, which make use of an independent
meaius of assessment, thus eliminating so far as possible the human element. However,
subjective methods are still used in clinical work, and it has been stated that no amount of
objective measurement of �vork capacity can be substituted for the subjective estimation of
au experienced surgeon (Milch 1945). The advantages of objective methods mentioned below,
however, warrant the use of quantitative measurement in conjunction with personal
impressions wherever possible.
The main disadvantage of a subjective method is that it is crude; there can he no fine
discrimination between different levels of muscle activity. For this reason it is of little or
ho use as a quantitative measure in fundamental or applied problems and has certain
disadvantages in clinical work. Impressions gained from observing muscle and joint action
may var�’ from time to time in the same observer and almost certainly do vary in different
observers. This may lead to incorrect conclusions, particularly if the rate of recovery in a
patient is slow ; and it is therefore essentialin the personal assessment of j oint and muscle function
for consecutive tests on any one patient to be made by the same observer. Discussing these
disadvantages, Lovett and Martin (1916b) stated: “ Impressions that electricity of one kind
or another, or rest, or exercise were beneficial have filled the literature ; unsupported assertions,
marvellous cures and fantastic treatments have too often been advanced on the slenderest
of grounds. ‘ ‘ A further disadvantage of the subjective method is that, since there are
iuecessarily few levels of function that can he distinguished, recorded improvement is slow
and the patient therefore derives little encouragement from the tests. An advantage of
considering only a few functional levels is that any one grade is wide enough to cover
incidental variations.
The niain advantages of a subjective method are that it is quick and requires no apparatus,
amid it can be used for any muscle or joint in individuals of any age. In the assessment of
muscle strength, the empirical index of function that it provides has the appearance of being
complete in itself and of requiring no comparison with a table of normal values as do
nunuerical results.
In coiutrast, the objective method may be time-consuming and relatively expensive and
yields results which require comparison with a standard. A further disadvantage that may arise
in ol)jective testing if an instrument is used for any length of time is that training or fatigue
h-nay affect thue results, The importance of these factors max’ possibly he assessed if measurements
are taken on the normal side for comparison.
‘I’hie main advaiutage of thie objective method is that it is reliable and, since measurememits
can he taken to the re(iuired degree of accuracy, the method has a wide application in studies
Oil iuormal individuals as well as in clinical work. Furthermore, small increases in muscle
strength or amplitude of movement can be demonstrated, which is important in providing
emucouragement to the patient as well as useful information for the clinician. Milch (1945)
pointed out that objective methods can provide readings which may lie accurately compared
withi further readings taken OD the same individual at different times. This is one of the
most important advantages of such methods, since the rate of recovers’, effect of treatment,
etc., can he assessed.
VOL. 37 B, NO. 3, AUGUST 1955
476 NANCY SALTER
From the following discussion it will be seen that the general trend is from simple
subjective methods to more accurate and complicated objective methods, and that, whereas
the former provide valuable information in routine clinical work, more reliable means have
been evolved for the accurate study of clinical and applied problems. One point of great
importance must be stressed : whether the tests made are subjective or objective, it is essential
that the results are recorded accurately and intelligibly.
METHODS OF MEASURING MUSCLE STRENGTH
Probably no single method of determining the strength of individual muscles or muscle
groups can be used for every circumstance in which such observations are required. Any
method, however, must allow reliable recordings to be quickly and easily made and must be
adaptable to the measurement of as many muscle groups as possible.
SUBJECTIVE ASSESSMENT OF MUSCLE FUNCTION
Clinical assessment of muscle strength was originally based on purely subjective methods.
The degree of muscle contraction was estimated manually and the muscles graded as normal,
partly paralysed or totally paralysed. The partly paralysed group included muscles ranging
from those that were nearly normal to those producing only a flicker of contraction. The
limitations of this method are obvious. This test was amplified by the estimation of the
amount of pressure that had to be applied to prevent a movement occurring. A certain degree
of objectivity in grading intermediate levels of activity was provided by adopting the following
system for muscles whose action is affected by gravity: None=no evidence of contractility;
trace�the muscle can be felt to tighten but cannot produce movement ; poor=movement
with gravity eliminated ; fair=movement against gravity ; good=movement against gravity
and some resistance ; norma1�movement against gravity and a greater resistance.
This scheme was first put forward by Lovett (1916), and since that time the different
levels of function have been given various numerical values. Legg (1936) and Nelson (1947)
used values from 0 (normal, or no paralysis) to 5 (complete paralysis), and Kendall and Kendall
(1949) modified the scheme slightly and graded muscle strength as a percentage of normal.
The grading most commonly used in Great Britain is that proposed by the Medical Research
Council in 1942 (\I.R.C. Memorandum No. 7)-namely, from 0 (no contraction) to 5 (normal).
This method is widely used in clinical work and has the advantage that it provides an
empirical measure of function and can be applied to any muscle group in the action of which
gravity is important, and in individuals of any age. However, the method is not clearly
defimied ; for example, a muscle may be able to overcome gravity plus some external resistance
through part of its range, but may not be able to complete the range of movement even
against gravity alone. Also, in practice, muscles are graded as 5 (normal) when they can
function adequately, although in fact they might compare unfavourably with unaffected
muscles. Thirdly, no standard techniques are adhered to. The following example is given
of two methods of grading the quadriceps muscle which were used in the same hospital:
I ) The knee is fully extended and the patient is required to maintain this position against
applied resistance attempting to flex the knee. 2) The knee is flexed to a right angle and
the patient attempts to extend it against resistance applied to oppose the movement.
Finally, the manual testing method is of no use for grading muscles whose action is either
not affected by gravity, such as the abdominal muscles, or which do not normally act alone,
such as the brachialis. Subjective methods in general are of little or no use for studying the
strength of normal muscles under different conditions.
OBJECTIVE METHODS
The methods ahud instruments to be described have been devised for the measurement
of strength of individual muscles or muscle groups and for the assessment of muscular
efficiency as a whole. Complicated ergometers for determination of energy expenditure
THE JOURNAL OF BONE AND JOINT SURGERY
METHODS OF MEASUREMENT OF MUSCLE AND JOINT FUNCTION 477
during prolonged periods of work involving large muscle groups, and instruments designed
primarily for specific experimental work are not described ; neither is the use of electro-
myography. The different methods to be described can be roughly classified into four
types : spring-balance methods, pressure systems, ergographic techniques and strain-gauge
methods.
Spring-balance methods-From the subjective methods for grading the functional activity
of partly paralysed muscles and muscle groups evolved the first objective method for the
large-scale determination of muscle strength. The subjective estimation of the amount of
resistance necessary to prevent a movement was replaced by a quantitative measure recorded
on a spring balance. One of the main differences between this and other tests of muscle
strength that have been proposed is that the part played by the subject is to resist a pull
rather than to exert his strength in active effort. This method was suggested and developed
by Lovett and Martin (1916a) and was the first simple technique for objective muscle testing
to be widely used in clinical work, although the principle of the method had been put forward
and a suitable instrument devised by Duchenne in 1863 (Bullard 1886). The method, employing
a simple spring balance, was able to show in muscles a slight degree of function not detectable
by other means, and more accurate quantitative values could be obtained to demonstrate
the course of recovery under different kinds of therapy. A sling adjusted round the limb
was attached to an accurate spring balance which measured the resistance applied in order
to prevent the limb moving from a standard position. The authors drew up a standard
technique for the major muscle groups and compiled a table of ‘ ‘ normal ‘ ‘ values for varying
ages (I�1artin 1921).
A similar spring-balance technique was used in an experimental study by Haxton (1944)
and in clinical work by Lewey, Kuhn and Juditski (1947). The latter workers validated the
use of their method b� comparing the strength of muscle groups expressed as a percentage
of normal, with the amplitude of the electromyogram obtained from maximal electrical
stimulation of the nerve also expressed as a percentage of normal. Schmier (1939) used a
modification of the method of Lovett and Martin incorporating an ink-writing pointer and
appliances for the stabilisation of the patient. Later he further modified and elaborated his
method, using a gravity lever scale similar to that used by %-Iayer and Greenberg (1942)
(Schmier 1945). These workers had criticised the technique of Lovett and Martin, stating
that a spring balance permits too much subjective variation in the hand of different examiners
and too great a range of body motion, with a consequent shift of the angle of force and a
variation in the strength of contracting muscles.
Clarke, Elkins, Martin and \Vakim (1950), in a study of normal muscles, measured the
amount of tension applied to a cable appropriately placed for specific movements as in
Lovett’s method, with a specially adapted and calibrated tensiometer. This instrument works
On the spring principle but is geared to be sensitive to contractions which are almost isometric.
Using springs, Rudd (1 951 ) developed a dynamometer which could be used for the measurement
of hand and finger strength and also for resistance exercises.
Two instruments commonly used in physical fitness tests as well as in clinical and
anthropological work are the hand-grip dynamometer and the oval-spring hand dynamometer.
Both these instruments incorporate a spring system but, as well as the disadvantage common
to such systems (see below) they are limited in use, for they can be used only to measure the
power of the hand grip and they are not adjustable for hands of different sizes. Regnier (1807)
(cited by Bullard 1886) was the first to devise a hand dynamometer of the oval-spring type.
A more compact instrument was developed by Mathieu and by Robert and Collin about 1870
(Bullard 1886) and has been extensively used with only minor modifications. The disadvantage
of using the grip dynamometer as an index of general strength was noted by Martin and
Rich (1918): the subject uses one of the most complex musculatures of the body and one
which receives much special training. The test therefore reflects conditions to an exceptional
VOL. 37 B, NO. 3, AUGUST 1955
478 NANCY SALTER
degree. However, the instruments are simple and easy to use and adaptable to a wide variety
of test circumstances.
The following disadvantages of a spring-balance system were noted by Wakim, Gersten,
Elkins and Martin ( 1 950) : a spring has a high internal resistance, a low range of sensitivity
and a chiamuging sensitivity. The fact that any spring can only work over a limited range of
sensitivity is probably the greatest disadvantage. A further disadvantage in any instrument
involving the compression or extension of a spring is that there is an increase in resistance
towards the end of the range of movement-that is, at the stage where the muscle is least
able to overcome it. A disadvantage of this and the following two methods of objective
muscle testing is that they cannot be used to test muscles which, while not completely
paralysed, are too weak to produce movement.
Pressure systems-Another group of instruments devised for the objective measurement
of muscle strength act by means of a system in which pressure changes occur. Most of these
�vere designed to measure the strength of hand grip. The earliest dynamometer of this type,
involving compression of a rubber bulb which causes coloured water to rise in a graduated
tube, was devised by Hamilton in 1875 (Bullard 1886). Fox (1917) described a similar
instrumeiut, which Amar called a ‘ ‘ dynamographic pear, ‘ ‘ in which the pressure change could
I)e read directly off a scale or recorded kymographically. Modifications of this simple apparatus
were used by Brahme (1936) and Schwab (1953), while Hellebrandt, Houtz and Kelso (1950)
developed a complex instrument incorporating a pressure system, changes in which activated
the lifting of weights.
I’arl)ell (1950) described a “ digit myometer “ made for measuring small gains in finger
strength after injury. Its main advantage appears to be in its simplicity ; the strength of the
finger is proportional to the increase in height of the column of mercury in an ordinary
sphygmomanometer, pressure being exerted on a lever by the finger and transmitted to the
cuff through a metal plate. A similar principle was used by Lewis, Pickering and Rothschild
(1931). Newman (1949) designed a myometer which could be used for most muscle groups.
The iiustrumemut consists of a pressure gauge and measures the resistance offered by a muscle
iii isometric contraction ; the reading on the gauge is proportional to the force necessary to
overconue the isometric contraction of the muscle under test. The gauge is set in a small
cylinder, froni one emid of ��‘hich extends a short shaft and a pressure-transmitting button.
A built-in luydraulic pressure converter transmits the linear force exerted on the button to
the pressure gauge.
Ergographic methods-These involve simple graphic representation of repeated muscular
exertions against a resistance provided usually by weights or springs. Instead of giving a
simugle value for maximum muscle strength at any one time, this method pro�’ides a record
shio�viiug thie rate of onset of fatigue �vith a load that may be initially maximal or sub-maximal.
Iii this way the emiduraiuce of a muscle or muscle group can be tested and the total work done
duritug a kno�vn 1)eriod of time can be calculated, thus indicating the functional capacity of
the muscle or muscle group (Molander and Weinmann 1942). A disadvantage of the method
is the length of time required for testing, which makes it impracticable for routine clinical
�vork. However, the imustrument can also be used for therapeutic exercise.
The’ original weight ergographi for recording finger contractions was devised by Mosso (1890)
for eXl)eriluue’hital studies on fatigue. The ergograph record was made with a pen carried by a
snuall llock which moved aloiug a hiorizontal track. On flexion of the finger, a string fastemied to
it by a leathier sling moves the block together with weights attached by a cord and passing over
a pulley. This principle has been used more recently by other workers in the clinical field
(Sotuntag I91 7, Simon amid Simonnet 1938, Kinard and Coleman 1946) as well as for experimental
studies (Martin 1921). One objection to the original ergograph is that as the moving part
moves imi an arc the angle of pull on the weight is continually changing, and therefore it is
difficult to calculate accurately the work done by the muscle. Hellebrandt and her co-workers
THE JOURNAL OF BONE AND JOINT SURGERY
METHODS OF MEASUREMENT OF MUSCLE AND JOINT FUNCTION 479
(Hellebrandt, Skowlund and Kelso 1948, and Hellebrandt, Kelso, Houtz and Eubank 1950)
have developed a complex modification of the apparatus incorporating a wheel to the
circumference of which is fastened a bar which is activated by the part under study. Thus
the angle at which the force is exerted remains constant throughout the range of movement.
In this instrument the total distance moved and the number of contractions are recorded
ihidependently of the kymograph tracing.
Simple weight lifting has been used in another way as a means of determining muscle
strength. Delorme (1945) advocated the use of repeated exertions with gradually increasing
loads to find thie “ one repetition maximum,” that is, the maximum load that can be lifted
through the full range of movement once only. A disadvantage of taking this as a strength
index appears to l)e that the patient may be fatigued by the time the maximal load is
determined (Zinovieff 1951). A further disadvantage of any weight-lifting technique is that
there may be dangers in moving heavy loads. If a subject suddenly becomes unable to lift
the load, a rapid movement in the reverse direction may cause muscle damage. There is also
the danger of falling weights.
Theoretically, the strength of muscles whichi are too weak to lift even the weight of the
limb agaimist gravity can be gauged by determining the weight necessary to assist the muscle
and enal)le movement to take place. In practice, however, it might be difficult to prevent the
muscle from playing a purely passive role.
Franz (1901) objected to the weight-lifting type of ergograph because it does not measure
isometric muscle contraction. He replaced the lifting of weights by the extension of a spring,
aiud this method was also used by Hough (1901) and Hall (1902) and more recently by Maison
and Broeker (1941) and Russell (1952). Some of the pressure-system instruments already
descrihed have been used as ergographs as well as for single tests (Brahme 1936, Hellebrandt,
Houtz and Kelso 1950, Schwab 1953).
Strain gauge methods-More recent methods for the accurate determination of muscle
strength involve the use of electrical strain gauges, the advantages of which are their low
inertia arid high degree and wide range of sensitivity. In such methods the applied muscle
force is allowed to deform metal bars or rings to which the gauges are attached. The
defornuation causes a change in the electrical resistance of the gauges which can he recorded
either omu a galvanometer or, after amplification, on an oscillograph or by pens. Ralston,
human, Strait and Shaffrath (1947) used a strain-gauge dynamometer for the measurement of
isometric contraction of human upper limb muscles in amputees. Strain gauges were attached,
two on the inner and two on the outer side of a ring set in a cable which was fixed at one end
and attached at the other end to the tendon of the muscle under study. Tension on the cable
produced changes in resistance of the gauges which were recorded oscillographically. Wakim
et oil. (195()) used this method on normal muscles, the cable being attached by a sling to the
extremity hieing tested. They found results similar to those obtained using the cable tensiometer
(Clarke et a!. 1950). Tuttle, Janney and Thompson (1950) used strain gauges in a hand-grip
dynamometer, and Darcus (1953) described a strain-gauge apparatus for the measurement of
rotary isometric torques about various joints.
The latter apparatus has recently been accepted for use in a Medical Research Council
imuvestigation on the effects of different exercise routines on the strength of weakened muscle
groups following poliomyelitis. This dynamometer appears to be highly suitable both for
research projects and for routine clinical testing. In clinical work, however, it would
supplement rather than replace the 0-5 muscle charting system.
MEASUREMENT OF THE AMPLITUDE OF JOINT MOVEMENT
The requirements of an “ideal” method for determining the amplitude of joint
movements are that it should allow the recording of accurate readings which are reproducible
by different observers, and that it should be applicable to all limb joints in people of different
\‘OL. 37 B, NC). 3, AUGUST 1955
H
48() NANCY SALTER
body dimensions. If the method demands an apparatus, this should be simple in construction
and use and should not interfere with joint movement. Most of the methods so far devised
fulfil only a few of these requirements.
The variety of methods described may be broadly divided into subjective assessments
and objective measurements with various types of apparatus.
SUBJECTIVE ASSESSMENT
Subjective assessment, which is dependent on visual approximation, was the first method
used amid is still commonly employed in clinical assessment of joint function. An adaptation
of the visual method was provided by Cleveland (1918). He used circular charts on to which
the limits of movement of the moving limb itself, estimated by visual means, were transferred.
In the centre of the charts were diagrams of the part concerned showing the axis of movement.
Clark ( 1 92 1 ) used a more refined modification of the visual method for the measurement
of rotary movements. He used a protractor scale arched over the distal end of the rotating
limb, withi the limb itself serving as an indicator.
OBJECTIVE METHODS
To allow measurements to be made accurately and to be independent of the observer,
numerous instruments have been devised during recent years. These have been termed
“ arthrometers ‘ ‘ (instruments for measuring joints), “ goniometers “ (instruments for
measuring angles) or “ fleximeters ‘ ‘ (instruments for measuring the degree of bending).
These may be classified roughly into three types : protractor and pendulum arthrometers
and independent methods.
The protractor arthrometer-This consists essentially of two rigid shafts intersecting at
a union allowing movement at right angles to their longitudinal axes. A protractor is fixed
to one of the shafts so that its centre corresponds with this union ; the other can move
independently of the protractor and acts as an indicator. The shaft to which the protractor
is attached is held or strapped parallel to the longitudinal axis of one part of the limb, the
union of the shafts corresponding to the axis of the movement being studied, and the other
is held against or strapped to the other part of the limb. The degree of movement can then
be measured directly from the protractor scale. Many modifications and improvement of
this basic pattern are reported in the literature (Camus and Faidherbe 1915, Manouvrier 1915,
Alquier 1916, Fox 1917, Dausset 1919, Albee and Gilliland 1920, Marble 1920, Rosen 1922,
Silver 1923, Conwell 1925, Cobe 1928, Parker 1929, West 1945, Dorinson and Wagner 1948).
All have been used to measure flexion and extension movements, particularly at the knee and
elbow.
Although the error inherent in the construction of these protractor arthrometers is
negligible, inaccuracies occur during their use, mainly from their faulty application. Such
faults include failure to ensure that the shafts of the instrument are parallel to the longitudinal
axes of the corresponding limb segments, and that the axis of the instrument coincides with
the axis of the joint. Wiechec and Krusen (1939), however, considered that with a moderate
amount of care in positioning, a simple form of protractor arthrometer, such as the one
described by Clark (1920), could attain as great a degree of accuracy as a much more
complicated instrument.
Wilson and Stasch (1945) have modified the technique of the measurement of joint
movement by the protractor method. They made double-exposure photographs of the
arthrometer and the limb, first with the joint in the position of maximal extension and then
with the joint in the position of maximal flexion. The amplitude of movement was read off
from the photograph, which also provided a permanent record. They also used this photographic
method without the arthrometer, determining the amplitude of movement on the developed
print by running two straight hines from the axis of the joint along the longitudinal axis of
THE JOURNAL OF BONE AND JOINT SURGERY
METHODS OF MEASUREMENT OF MUSCLE ANI) JOINT FUNCTION 481
the moval)le part of the limb in the two positions, and measuring the angle with a protractor.
This latter method was found to be particularly useful in observations on the wrist, metacarpo-
phalangeal and interphalangeal joints. These methods do not seem to have any particular
advantage over other protractor methods, apart from providing a permanent graphic record,
and this is outweighed by the increased complexity of the procedure.
The pendulum arthrometer-This type of instrument consists of a circular scale, to the
centre of which is attached a pointer weighted at one end so that it remains vertical while
the scale rotates around it. The scale of the instrument is attached to the movable limb, and
during movement of the joint the scale moves around the stationary pointer, thus registering
the amplitude of movement. Falconer (cited by Fox and van Breeman 1934) appears to have
been the first to devise such an instrument for measuring the amplitude of movement of the
knee and elbow joints. Similar types for general use have been described by Glanville and
Kreezer (1937) and by Hand (1938).
Independent methods-A number of methods which are to a certain extent independent
of the subject on whom measurements are taken have been used to study the amplitude of
movements. With such methods there is no possibility of the apparatus interfering with
normal movement. Two methods are suitable for clinical use.
Wilmer and Elkins (1947), dissatisfied with other methods of measurement of joint
movement, devised an arthrometer embodying optical principles. This instrument consists
of a transl)arent circular scale attached to a concave reducing lens. On looking at the limb
through the lens and scale, the reduced image of the limb appears on the scale. The skin is
marked to indicate the axis of movement of the joint and define the longitudinal axes of the
bones forming the joint. The limits of movement are marked on the scale by movable pointers.
This method is claimed to have the advantages of simplicity, lack of interference with the
movement, and adaptability.
Recently Zankel (1951) has described another method which he terms “ photogoniometry.�’
A protractor scale is projected on to the limb of the subject so that the centre of the scale is
focused on the axis of the joint. The amplitude of movement is measured from the projected
radii of the protractor scale, which coincide with the longitudinal axes of the bones articulating
at the joint under consideration. Points may be marked on the skin to facilitate the readings.
He claims that it is an improvement on previous methods because it is more accurate alid
more susceptible to duplication by different observers, although he gives no evidence for this.
He also contends that this method lends itself better to teaching because many different
observers can study the same joint measurements concurrently.
Other methods that have been put forward involve radiographic and cinematographic
analyses of movement, but these were not intended for routine clinical assessment of amplitude.
Bakke (1931) studied movements of the vertebral column by serial radiographs, and Harris
and Joseph (1949) used radiography to measure the degree of extension of the metacarpo-
phalangeal and interphalangeal joints of the thumb. Weddell and Darcus (1947) and Darcus
(1948) studied the amplitude of composite movements of the neck and spine by the anah’sis
of a cinC-film record. CinCradiography has been used to study movements of the lower limb
(Barnett and Napier 1952) and may be of value in determining the amplitude of complex
movements.
A review of methods used to measure the amplitude of rotatory movements of thie
upper limb was given by Darcus and Salter (1953).
In conclusion it can be seen that, despite the efforts made by numerous workers to devise
an arthrometer fulfilling all or nearly all of the requirements, there is still no widely accepted
method. If used with sufficient care, it would seem that the simple protractor type of
arthrometer is the most generally useful, since it provides adequate information on most
joints. Where more precise measurements are necessary, a radiographic method is to be
recommended.
VOL. 37 B, NO. 3, AUGUST 1955
482 NANCY SALTER
FACTORS TO BE OBSERVED IN THE USE OF APPARATUS AND METHODS DESCRIBED
In connection with the use of both objective and subjective tests for the measurement
of muscle strength and the amplitude of free movements, there are certain factors which merit
attention and consideration. The following points are put forward to illustrate the confusion
that is liable to arise when, for example, different terminologies are used, and to provoke
suggestions and perhaps some measure of standardisation which would be of benefit in this
field of study. It is hoped also that the discussion will demonstrate the importance of
indicating iii case records and published results the exact technique and procedure of a test
so that adequate data are available for those wishing to repeat the study or to use the results
for comparative purposes. The factors considered here, although there may be others equally
important, are posture, test procedure, standards for comparison, nomenclature and normal
variability.
Posture-In the assessmeiut of both muscle strength and the amplitude of joint movements,
it is necessary to standardise, or at least to specify, the posture of the body and the precise
1)OsitiOIi of the joints under study as well as those of associated joints. It is essential that
these J)ositiolis should be the same on successive occasions on which measurements are made
in order to be able to compare readings taken from the same part in the same individual at
different times. The position of the body-for example, sitting or lying down-is important
froni the point of view of comfort and relaxation of the patient amid in so far as it affects the
l)ositiOIi of the limbs, Fox (1917), stressing this point in the measurement of the amplitudeof nuovement, gave as an example the increased flexion of the hip allowed by flexion of the
knee’ �vhuich reduces the ‘ ‘ ligamentous ‘ ‘ action of the hamstring muscles. Similarly the
amplitude of dorsiflexioiu of the foot is restricted by extension of the knee due to the
‘ ‘ liganientous ‘ ‘ actiomi of the gastrocnemius and the amplitude of flexion of the metacarpo-
plialamugeal joints of the hand by flexion of the wrist due to the ligamentous action of the
long extensors. Thue results of Cobe (1928) show also that the position of the radio-ulnar
joints influences the amplitude of abduction and adduction of the wrist.
Experimental studies on the effect of joint position on muscle force (Franke 1920,
Hamusen and Lindhuard 1923, Hvorslev 1928, Garry 1930, Muller 1935, Haxton 1945, 1)ern,
Levemue and Blair 1947, Hugh-jones 1947, Clarke, Elkins, Martin and Wakim 1950, Wakim,
(;ersto’ii, Elkins and Martin 1950, \Vilkie 1950, Darcus 1951, Salter and Darcus 1952) show
how importamit is the consideration of individual joint position in the assessment of muscle
strengtlu, �vhether this assessment is subjective or objective, and for clinical or fundamental
�vork. Posture is also important in the use of the simple and commonly used spring-balance
dynamometers, and results of different workers may be of little value for comparison unless
standard positions are adopted. This is important, for example, in the ‘ ‘ lumbar pull ‘ ‘ test
which is widely used to obtain an index of general strength. In this test the subject holds the
dvnamometer handle and pulls vertically upwards, keeping his arms and legs straight.
Vernon (I 924) had shown that there is an optimum height for the handle below or above
which the force diminishes considerably, and after experimental investigation Bedford and
Warner (1937) found that the greatest pull was exerted when the handle was at the height
of the subject’s finger tip.
No standard posture has been advocated for the use of the hand-grip dynamometer.
The subject is usually instructed to hold the instrument where he feels that he can exert
his greatest force. Fisher and Birren (1946) found that supporting the instrument with the
other hand affected the results considerably. It is possible that using a standard posture for
the test may also affect the results.
In all muscle testing, both for strength and amplitude of movements, it is necessary to
specify the part played by the force of gravity in assisting or resisting movement. It is also
necessary to prevent, or at least to be able to detect, substitution or “ trick” movements so
that one is sure what muscle or muscle groups are acting.
THE JOURNAL OF BONE AND JOINT SURGERY
METHODS OF MEASUREMENT OF MUSCLE AND JOINT FUNCTION 483
Most accurate results of objective muscle testing would be obtained in the positions of
associated joints which normally allow the greatest force to be exerted. This information is
available for certain muscle groups (Clarke et al. 1950, Darcus 1951).
Test procedure-In all muscle tests the question arises of how many observations should
1)e made on a single occasion. More than one reading is usually taken to avoid chance
inaccuracy. Hellebrandt, Parrish and Houtz (1947) showed that an increase in the strength
of muscles, which had undoubtedly occurred, was not reflected in tests involving only one
reading on a spring dynamometer. They attributed this to the many uncontrollable variables
which may influence an isolated observation. However, a study of the reliability of the single
muscle test was made by Duvall, Houtz and Hellebrandt (1947) and they found that a single
reading is as reliable as the best of three or the best of ten. For clinical measurements of
joint amplitude three readings are usually taken. Common methods in the measurement of
muscle strength are to take the mean of two or three observations or the best of three ; these
I)rol)a1)1Y evolved as being the most convenient and the least time-consuming. Doubting the
reliability of the “ best of three “ method, Fisher and Birren (1946) tested the following
procedure with a hand-grip dynamometer : scores were taken at three-second intervals withthe subject starting at 27 kilograms and increasing by increments of 3 kilograms until his limit
was reached. The load increment and the time interval were chosen arbitrarily. Re-test
J)erformance showed this to be a reliable procedure and the authors gave it the followingadvantages : it allows a ‘ ‘ warming up ‘ ‘ to take place, and the cumulative fatigue ensures
the final effort being made under a standard stress. A similar technique was used by Delorme
(1945) in clinical work ; his subjects worked with a gradually increasing load until a standard
number of contractions could not be attained. In considering the number of readings to be
taken, the relative importance of fatigue and ‘ ‘ warming up ‘ ‘ must be noted. The best time
interval between successive recordings must be determined, and we have found that five
maximal voluntary exertions spaced at intervals of one minute produce little or no fatigue.
The speed of making a single exertion-that is, the time over which the maximal force is
huilt up in the muscle, must also be considered. Studies carried out by Hill (1922) indicate
that the work done in a single maximal effort increases with increasing time taken to build
up the contraction-that is, with decreasing speed of contraction. This is due to the fact
that the more rapidly a muscle shortens the more potential energy is used up in overcoming
the viscous resistance of the muscle. However, there is an optimal economical speed of
working related to maximal efficiency.
In ergographic techniques the question arises of the rate at which maximal contractions
should 1)e made. This is determined partly by the movement under consideration and partly
by whether or not the muscles are normal. Common time intervals are one, two or three
seconds. One bout to complete exhaustion may be recorded or a standard number of
contractions may be performed and repeated after intervals of known duration. Foltz, Ivy
and Barborka (1942) recommended the use of double work periods, each to exhaustion, with
a standard rest pause between. The advantage of this method is that, as well as the total
work done, the percentage recovery is obtained.
Standards for comparison-In clinical work some standard for comparison is required
when measuring either the force in a paretic muscle or the limited amplitude of movement
of a joint. As a standard the corresponding measurement taken on the contralateral limb
may be used, if this is normal, or reference may be made to a relevant table of ‘ ‘ normal”
values. So-called normal values for the amplitude of various movements are to be found in
the literature (Clark 1920, Gilhiland 1921, Rosen 1922, Silver 1923, Frescoln h929, Sinelnikoff
and Grigorowitsch 1931, Couhter and Molander 1935, Glanville and Kreezer 1937, Dorinson
and Wagner 1948). Gillihand’s values were obtained from” 100 observations on male students.”
However he does not specify the number of observations on each joint. Glanville and
Kreezer (1937) studied a number of joints in ten subjects, and the values given by Sinelnikoff
\‘C)L. 37 B, NO. 3, AUGUST 1955
484 NANCY SALTER
and Grigorowitsch were obtained from 396 male and 1 18 female subjects. The other workers
mentioned give no indication of how their values were obtained or of the size and characteristics
of the group of subjects. Such ‘ ‘ average ‘ ‘ or ‘ ‘ normal ‘ ‘ values are of limited use, particularly
if thue range of values which can be considered normal is not given, and, in fact, the values
givelu I)y different authors for the same movement differ from each other considerably (Darcus
and Salter 1953, Salter and Darcus 1953). A more accurate yardstick for the normal amplitude
of movement of a joint is probably the contralateral limb, when this is available for study.
That there is little difference between the normal amplitudes of certain movements on the
right and left sides has been stated by Gilliland (1921). Cobe (1928) and Hewitt (1928),
studying the amplitude of movements at the wrist, found differences between right and left
sides of over 200 per cent in some cases, but bigger differences were found between individuals,
and variations of over 100 per cent were recorded in taking the same measurement on the
same individual on different days, suggesting faulty apparatus or technique. Patrick (1946)
stated that the range of pronation and supination in both forearms of a single individual is
identical. Darcus and Salter (1953) found that the mean difference between the amplitudes
on the right and left sides in eight normal subjects was 3 per cent, in contrast with a difference
of 20 per cent between subjects.
Considering the measurement of muscle force, Lewey, Kuhn and Juditski (1947) put
forward the following objections to using the contralateral side as a standard for comparison:
I ) the time of the examination is doubled ; 2) there may be a compensatory increase in muscle
strength on the normal side ; and 3) values on the right side may be higher than on the left
side in right-handed subjects. Although the time of testing is doubled, readings taken on
the normal side are an advantage in that they reflect normal variations (Darcus 1951, Salter
and Darcus 1952). Variations due to psychological, physical and environmental factors can
be taken into account in assessing the affected side, as can training and fatigue. At present,
comprehensive tables of normal values suitable for comparison under particular test conditions
do not exist. Individual variation is so wide and dependent on so many factors (age, sex,
physical characteristics, occupation, etc.) that a great deal of work would be necessary to
compile adequate information, Using such information, one can only say that a patient falls
within the limits of normal variation for his age and sex group or is approximately x per cent
of “normal. “ Lewey et al. (1947) gave a table of normal values for twenty-three muscle groups
compiled from observations on a random sample of hospital patients showing no nerve or
muscle weakness. They give mean values and standard deviations but omit to give the
number of subjects used and their ages, merely stating that the group includes manual and
clerical workers and both right- and left-handed persons. They claim to have obtained more
satisfactory results using this table than by using the contralateral limb as a yardstick,
despite the fact that overall variation between different individuals appears to be much
greater than that between right and left sides in a single individual (Provins 1955, Provins
and Salter 1955).
Nomenclature-To be of value, records must be uniform and if possible self-explanatory.
Difficulties involved in assessing early work on the measurement of j oint amplitude included the
wide disagreement in nomenclature, which made results liable to misinterpretation. Each author
adopted his own nomenclature, often without specifying what he meant by the terms used.
Most workers appear to agree that all measurements should begin from a neutral point,
hut there is considerable difference of opinion as to where in the whole arc of movement
this should be. Usually it is placed at an assumed mid-point or at one or other end of this arc.
If the neutral point is at the assumed mid-point, differences are found in the angular value
given to this point and in the system of denoting degrees of movement on either side of it.
The difficulties of interpretation arising from this lack of consistency are indicated by the
figures given 1w different authors for the amplitude of flexion and extension of the wrist
(Table I).
THE JOURNAL OF BONE AND JOINT SURGERY
METHODS OF �IEASUREMENT OF MUSCLE AND JOINT FUNCTION 483
The meaning of these figures is apparent only when their graphic representation is
studied (Fig. 1).
Further confusion arises through the different methods of describing a movement. For
instance, iii joints allowing extension beyond the point at which the bones are in line, the
movement beyond this point is sometimes referred to as hyperextension. Abduction and
adduction of the wrist are also described as radial and ulnar deviation or flexion. �-Iovements
of the shoulder joint are either defined in relation to the main axes of the body or to the axes
of the scapula (Johnston 1937).
The first attempt to clarify the nomenclature was made by Clark (1920). He suggested
that the angle included between the bones on either side of the joint should be used to express
the limits of motion or angle of deformity ; thus complete extension of the knee or elbow
would he I 80 degrees. However, modification of and additions to this scheme for the
measurement of abduction, adduction and rotation made it unnecessarily complicated, and
it was not generally accepted. Rosen (1922) believed that measurements from a neutral
point were not satisfactory and suggested that the angle subtended at the joint axis by the
two extremes of excursion of the joint was more suitable. Silver (1923) proposed that
‘ ‘ mensuration of the degrees of motion in joints must start from a zero plane. The zero plane
should be considered to be an extension of the plane of the long bone or bones immediatel��
TABLE I
FIGURES OF DIFFERENT AUTHORS FOR AMPLITUDE OF
FLEXION AND EXTENSION AT THE WRIST
� Flexion Extension H vperextension
(degrees) (degrees) (degrees)
Clark (1920) 100 � 180 120
Robinson (1921) 110 � 115
Rosen (1922) � 135 225*
Silver (1923) 70 � 70
* These figures were not intended to represent the complete amplitude.
proximal to the joint in question.” This method had previously been used by Robinson (1921).
Cave amid Roberts (1936) merely stated that “ all movements should be measured by degrees
from a mieutral point or zero which must be defined. “ West (1945) measured all movements
on a circular scale graduated from 0-360 degrees, using the position of the joints in the
anatomical position as a neutral position. Flexion was measured towards 0 degrees and
extension towards 360 degrees.
Dorinson and Wagner (1948) put forward another suggestion in order to “ facilitate
recording and obviate any misinterpretation. ‘ ‘ They state that ‘ ‘ if, for an example, a patient
has a 10 degrees’ flexion contracture with a range restricted to 40 degrees, this would be
obvious by the recording of ‘ extension-flexion ‘ 170-130 degrees.” So far as is known, this
suggested method has not been applied to any extent.
It is evident that there is still as much disagreement as ever concerning nomenclature,
and that until a universal method is adopted each author must specify exactly which angle
he is in fact measuring and which neutral point he is using. To avoid ambiguity, it is stressed
that graphic representation of the results should supplement the written description (Cleveland
1918, Nutter 1919, Parker 1929). It is also important to state whether the movements
measured are active or passive. The results of Glanville and Kreezer (1937) indicate that the
difference between the amplitudes of active and passive movements in normal individuals
\‘OL, 37 B, NC). 3, AUGUST 1955
Flex,
Fore arm
486 NANCY SALTER
TilE JOURNAL OF BONE ANI) JOINT SURGERY
nuay he 10 per cent or more, and it is felt that this question deserves further investigations.
in climuical con(litiOIus th( difference max’ he much larger and, while this is well kmuown, no
objective clituical study on thue slll)ject has been found.
Similar confusion is less likely to result from work on the measurement of muuuscle force,
providing thuat thie units and thue point of application of force are clearly defined, so that only
strictly conuparable valuo’s are in fact compared. Static activity cannot l)e conipare’d directly
CLARK Hypcr-cxt.
Ext. #{163}82-
ARC OF MOVEMENT�l4O#{176}
ROBINSON
ARC OF MOVEMENT.l3S#{176}Flex
ROSEN Ext\��
/�ARC OF MOVEMENT�9OFlex.
Forearm
SILVER
ARC OF MOVEMENT= (40#{176}
4
For co rm
FIG. 1
Dmagramn to mnuhmcate’ the nomenclatumre and systems ulsed b�’ varioums aumthors
to demiote’ the degree of flexmon and e’xtension of the wrist.
�vithu (lyIiaIuuiC work lx’cause thue’ units are different. Starr (1951), huowever, huas devised a
niiathiemiuatical Iflethiod of expressing the t�vo components in experimental weight lifting in
the’ SaITI(’ units.
imu rnaiuual muscle testimug in Great Britain muscles are graded numerically from (1) to 3,
0 imudicating no evidt’nce of comitractility and 5 a normal muscle (Lovett 1916, Medical Research
(‘ouncil Memorandum No, 7,1942). It was confusing to find in two American papers the same
METHODS OF MEASUREMENT OF MUSCLE AND JOINT FUNCTION 487
functional levels given different numerical values (Legg 1936, Nelson 1947). These numerical
values may l)e misleading as they bear no relationship to absolute levels of muscle strength.
For examj)le, a muscle of grade 4 is not twice as strong as a grade 2 muscle.
In order to avoid difficulty in the interpretation of results, it is stressed that the exact
procedure should in the first instance be described.
Variability-Considerable variation is found in measurements of muscle force and of
amplitude taken on different occasions in the same individual (Darcus 1951, Salter and Darcus
1953). Such variations will occur in injured as well as normal individuals, and the clinician
will wish to know how big a change in muscle force or amplitude must be before it reaches
practical significance. A dramatic and maintained improvement offers no problem, and small
changes which are insignificant in themselves become significant when they are reflected in
the trend of a series of tests made on successive occasions. However, in order to increase
the accuracy of assessing any improvement, normal variability should be minimised so far
as possible.
Ofenvironmental factors, temperatureand possiblyhumidity and barometricpressure may
affect muscle strength. Probably temperature is the most critical, though it can most easily
he controlled. Lovett and �t1artin (1916b) noticed that during “ the great heat of September
I 91 5 ‘ ‘ there was a decrease in muscle strength in several persons whose normal values were
known. The temperature of limbs is important in testing muscle strength, particularly after
poliomyelitis. Lombard (1892) concluded from his experiments that the time of day and
barometric pressure influences muscle strength, and the latter has also been suggested by
Fischer (1947). These factors have not been proved to have an effect, but the time at which
tests are made could perhaps be standardised for each subject.
Distractions to the patient and to the observer should be avoided, concentration on the
test maintained and motivation standardised as far as possible. In clinical work the incentive
is usually high, particularly if the patient is allowed to watch the readings being taken and
to follow the results of the test (Albee and Gilliland 1920). The willingness to exert a maximal
effort when this is required may markedly affect the results, and there is no simple objective
method of distinguishing between the physiological and psychological end points of effort.
The interest and co-operation of the subject should be sought. This is particularly important
in experimental work with normal subjects, because boredom and antagonism would give
inaccurate results and wide variation within groups of readings. Individuals being tested
should be instructed to relax ‘ ‘ mentally ‘ ‘ as well as physically, for apprehension or anxiety
may affect the degree of muscle tension (Jacobson 1946, Barlow 1947, Lundervold 1952).
The necessity for rigid control of environmental, physiological and psychological factors is
emphasised by Taylor and Brozek (1944). Factors which could be to some extent controlled
in hospital patients, but which are difficult if not impossible to control in other subjects,
are the amount of rest and the amount and type of activity and exercise. These factors may
affect readings taken at long intervals or over extended periods. Pain may limit the degree
of muscle contraction in clinical cases and must be taken into account when relevant. If one
side of the body is not affected by disease or injury, control tests on this side might help in
distinguishing between the effect of environmental factors and factors acting on the affected
side alone. Similarly, contralateral tests might assist in the distinction between a training
effect in the remaining innervated fibres in a paretic muscle and improvement perhaps due to
functional re-innervation of previously denervated fibres.
In summing up this discussion of the importance of test procedure, posture, nomenclature,
standards for comparison and normal variability in the measurement of muscle and joint
function, it can be seen that a consideration of these factors is essential if the work carried
out is to be as useful as possible. Although it is impossible to apply rigid techniques, because
modifications are required to meet individual needs, it is considered that there is room for
sonic standardisation, particularly of nomenclature. Until there is a generally accepted system
VOL. 37 B, NO. 3, AUGUST 1955
488 NANCY SALTER
of denoting the angular position of joints it is necessary for workers to describe accurately
the system they use as well as specifying details of posture and test procedure. It is
appreciated that time is valuable in routine hospital work, but the little extra time required
to specify essential details of the procedure would be well spent. If this is done, future
I)resehitation of data will he clearer and more useful than it has been before.
SUMMARY
I . The importance of accurate methods of measuring the strengthi of muscles and the
anuplitude of joint movements in man, both in clinical fields and as criteria of normal function,
is discussed.
2. The advantages and disadvantages of subjective and objective methods are reviewed.
3. The main types of apparatus used for the assessment of muscle strength in both normal
and clinical conditions are described. A dynamometer of the strain-gauge type is recommended.
4. Methods of measuring the amplitude of movements in man are also described. The
protractor type arthirometer is thought to be the most suitable for routine clinical work, but
for research purposes a radiographic method may be preferable.
5. The following factors, which must be considered if the measurements taken are to be of
greatest use, are discussed : posture, test procedure, standards for comparison, nomenclature
and normal variability.
REFERENCES
ALBEE, F. H., and GILLILAND, A. H. (1920) : Metrotherapy, or the Measure of Voluntary Movement. Journal
of the American Medical Association, 75, 983.
ALQUIER, L. (1916): Un goniom#{233}tre pr#{233}cis. Revue Neurologique, 23 (2), 515.BANKE, S. N. (1931) : Rontgenohogische Beobachtungen uber die Bewegungen der \Vmrbels#{228}ule. Acta
Radiologica. Supplememitumm XIII.
BARLOW, W, (1947) : Anxiety amid Muscle Tension. British Journal of Physical Medicine, N.S. 10, 81.BARNETT, C. H., and NAPIER, J. H. (1952) : The Axis of Rotation at the Ankle Joint in Man. Its Influence
upon the Form of the Taluis and the Mobility of the Fibula. Journal of Anatomy, 86, 1.
I3EuF0RD, ‘F., and VI’ARNER, C. (�. (1937): StrengthTests: Observations on the Effects of Postuire on Strength
of Pumll. Lancet, ii, 1,328.
BRAHME, L. (1936) : A Newly Constructed Energodynamometer and its Clinical Use. Acta Medica
Scanehinavica, 89, 268.BULLARD, W. N. (1886) : In A Reference Handbook of the Medical Sciences. Edited by A. H. Buck, 2, 544.
New York: William Wood & Co.
CAMUS, J., and FAIDHERBE, P. (1915) : Mesures des angles articulaires et des muscles situ#{233}sau-dessus et
aum-dessous des articulations. Mesure de Ia pronation et de ha supination. Comptes rendus hebdomadaires
des Seances et M#{233}moires de ha Soci#{233}t#{233}de Biologic, 78, 291.
CAVE, E. F., afl(l ROBERTS, S. M. (1936) : A Method for Measuring and Recording Joint Function. Journal
of Bone and Joint Surgery, 18, 455.
CLARK, \V. A. (1920) : A System of Joint Measurements. Journal of Orthopaedic Surgery, N.S. 2, 687.
CLARK, \V. A. (1921) : A Protractor for Measuring Rotation of Joints. Joumrnal of Orthopaedic Surgery,
N.S. 3, 154.
CLARKE, H. H., ELKIN5, E. C., MARTIN, G. M., and WAKIM, K. G. (1950): Relationship between Body
Position and the Application of Muscle Power to Movements of Joints. Archives of Physical Medicine, 31, 81.CLEVELAND, D. E. H. (1918) : Diagrams for Showing Limitation of Movements through Joints. Canadian
Medical Association Joumrnal, 8, 1,070.
COBE, H. M. (1928): The Range of Active Motion at the Wrist of White Adults. Journal of Bone and Joint
Surgery, 10, 763.CONWELL, H. E. (1925): Flexo-Extensometer. Surgery, Gynecology and Obstetrics, 40, 710.
COULTER, J. S., and MOLANDER, C. 0. (1935): Therapeutic Exercise. Journal of the American Medical
Association, 104, 118, 213.DARCUS, H. D. (1948): The Anatomical Principles Related to Sighting. Report No. R.N.P. 48/474, Prepared
for the Royal Naval Personnel Research Committee.
DARCUS, H. D. (1951): The Maximum Torques Developed in Pronation and Supination of the Right Hand.
J oumrnah of Anatomy, 85, 55.
THE JOURNAL OF BONE AND JOINT SURGERY
METHODS OF MEASUREMENT OF MUSCLE AND JOINT FUNCTION 489
DARcC’s, H. D. (1953) : A Strain-Gauge Dynamometer for Measuring the Strength of Muscle Contraction
and for Re-educating Muscles. Annals of Physical Medicine, 1, 163.
DARCUS, H. D., and SALTER, N. (1953) : The Amplitude of Pronation and Supination with the Elbow Flexed
to a Right Angle. Journal of Anatomy, 87, 169.
I)A�’SSET, H. (1919) : Enregistreur de travail et d’amphitude articumlaire, son utihisation pour he diagnostic
(les imlipotemices et poumr le dosage et he contr#{244}le des traitements m#{233}canoth#{233}rapie. Bulletin de I’Acad#{233}mie
de M#{233}clecmne, 3, s#{233}r.81, 819,
DELORME, T. L. (1945) : Restoration of Muscle Power by Heavy-Resistance Exercises. Journal of Bone
and Jomnt Suirgery, 27, 645.
DERN, H. J., LEVENE, J. M., and BLAIR, H. A. (1947) : Forces Exerted at Different Velocities in Human
Arm Movements. American Journal of Physiology, 151, 415,
I)0RINS0N, S. M., and \VAGNER, ItI. L. (1948) : An Exact Technic for Clinically Measuring and Recording
Joint Motion. Archives of Ph�’sical Medicine, 29, 468.
I)UVALL, E. N. (1948) : Tests and Measurements in Physical Medicine. Archives of Physical Medicine, 29, 202.
I)UVALL, E. N., HOUTZ, S. J., and HELLEBRANDT, F. A. (1947): Reliability ofa Single Effort Muscle Test.
Archives of Physical Medicine, 28, 213.
FISCHER, E. (1947) : Muscle Strength and the \Veather. Archives of Physical Medicine, 28, 295.
FISHER, M. B., and BIRREN, J, E. (1946) : Standardization of a Test of Hand Strength. Jotmrnal of Applied
Psychology, 30, 380.
FOLTZ, E., IVY, A. C., and BARBORKA, C. J. (1942) : The Use of Double Work Periods in the Study of Fatigume
and the Influence of Caffeine on Recovery. American Journal of Physiology, 136, 79.
Fox, H. F. (1917) : I)emonstration of the Mensuration Apparatus in Use at the Red Cross Clinic for the
Physical Treatment of Officers. Proceedings of the Royal Society of Medicine (Section of Balneology and
Climatology), 10, 63.
Fox, Ft. F., and BREEMEN, J. van (1934) : Chronic Rheumatism, Causation and Treatment, pp. 327-331.
London: J. & A. Churchill Ltd.
FRANKE, F. (1920) : Die Kraftkurve menschhicher Muskeln bei wihhkurlicher Innervation und die Frage der
al)soluiten Muskelkraft. Pflugers Archiv f#{252}rdie Gesamte Physiologic, 184, 300.
FRANZ, S. I. (1901) : On the Methods of Estimating the Force of Voluntary Muscular Contraction and on
Fatigume. American Journal of Physiology, 4, 348.
FRESCOLN, L. D. (1929) : Range of Bodily Movements. Medical Times, New York, 57, 197.
GARRY, R. C. (1930) : The Factors Determining the i\Iost Effective Push or Pull which can be Exerted by a
Humnan Being on a Straight Lever Moving in a Vertical Plane. Arbeitsphysiohogie, 3, 330.
GILLILAND, A. H. (1921) : Norms for Amplitude of Voluntary Movement. Journal of the American Medical
Association, 77, 1,357.
GLANVILLE, A. D., and NREEZER, G. (1937) : The Maximum Amplitude and Velocity of Joint Movements
in Normal Male Human Adults. Human Biology, 9, 197.
HALL, \\. S. (1902) : A New Form of Ergograph. American Journal of Physiology, 6, xxiii.
HAND, J. G. (1938) : A Compact Pendulum Arthrometer. Journal of Bone and Joint Surgery, 20, 494.
HANSEN, T. E., and LINDHARD, J, (1923) : On the Maximum Work of Human Muscles Especially the Flexors
of the Elbow. Journal of Physiology, 57, 287.
HARRIS, H., and JOSEPH, J, (1949) : Variation in Extension of the Metacarpo-Phahangeal and Interphahangeal
Joints of the Thumb. Journal of Bone and Joint Surgery, 31-B, 547.
HAXTON, H. A. (1944) : Absolute Muscle Force in the Ankle Flexors of Man. Journal of Physiology, 103, 267.HAXTON, H. (1945) : The Function of the Patella and the Effects of its Excision. Surgery, Gynecology and
Obstetrics, 80, 389.
HELLEBRANDT, F. A., PARRISH, A. M., and HOUTZ, S. J. (1947) : Cross Education. Archives of Physical
1\ledicine, 28, 76.
HELLEBRANDT, F. A., SKOWLUND, H. V., and KELSO, L. E. A. (1948) : New Devices for Disability Evaluation.
Archives of Physical Medicine, 29, 21.
HELLEBRANDT, F. A., HOUTZ, S. J., and KEL5O, L. E. A. (1950): New Devices for Disability Evaluation.
3. The Grip Ergograph. Archives of Physical Medicine, 31, 207.
HELLEBRANDT, F. A., KEL50, L. E. A., HOUTZ, S. J., and EUBANK, R. N. (1950): The Thumb Ergograph.
Archives of Physical Medicine, 31, 201.
HEWITT, D. (1928): Range of Active Motion at the \Vrist of Women. Journal of Bone and Joint Surger�’,
10, 775.
HILL, A. \‘. (1922): The Maximum Work and Mechanical Efficiency of Human Muscles and their Most
Economical Speed. Journal of Physiology, 56, 19.
HOUGH, T. (1901): Ergographic Studies in Neuro-Muscular Fatigue. American Journal of Physiology, 5,240.
HUGH-JONES, P. (1947): The Effect of Limb Position in Seated Subjects on their Ability to Utilize the
Maximtmm Contractile Force of the Limb Muscles, Journal of Physiology, 105, 332.
\‘OL. 37 B, NO. 3, AUGUST 1955
490 NANCY SALTER
HVORSLEV, C. M. (1928) : Stumdien fiber die Bewegungen der Schulter. Skandinavisches Archiv f#{252}rPhysiologic,
53, 1.
JACOBSON, E. (1946) : Electrical Measurements of Mental Activities in Man. Transactions of the New York
Academy of Science, 8, 272.
JOHNSTON, T. B. (1937) : The Movements of the Shoulder-Joint. British Journal of Surgery, 25, 252.
KENDALL, H. 0., and KENDALL, F. P. (1949): Muscles: Testing and Function. Baltimore: The \Vilhiams &
\Vilkins Company.
KINARD, F. \V., and COLEMAN, C. 1). (1946): A Modification of the Ergograph. Science, 103, 731.
LEGG, A. T. (1936) : The Early Treatment of Pohiomvehitis and the Importance of Physical Therapy.
Journal of the American Medical Association, 107, 633.
LEwEY, F. H., KUHN, \V. G., Jun., and JUDITSKI, J. T. (1947): A Standardized Method for Assessing the
Strength of Hand and Foot Mumsches. Surgery, Gynecology and Obstetrics, 85, 785.
LEwIS, T., I�ICKERING, G. \V., and ROTHSCHILD, P. (1931) : Observations upon Muscular Pain in Intermittent
Claumdication. Heart, 15, 359.
LOMBARD, \V. P. (1892) : Some of the Influences which Affect the Power of Voluntary Muscular Contractions.
Journal of Physiology, 13, 1.
LOVETT, R. \V. (1916) : The Treatment of Infantile Paralysis. Philadelphia: Blakiston’s Son & Co.
LOVETT, Ft. W., and MARTIN, E. G. (1916a) : The Spring Balance Muscle Test. American Journal of
Orthopaechic Sumrgery, 14, 415.
LOVETT, Ft. \V., and I’�lARTmN, E. G. (191Gb) : Certain Aspects of Infantile Paralysis with a Description of a
Method of Mumscle Testing. Joumrnah of the American Medical Association, 66, 729.
LUNDERVOLD, A. (1952) : An Ehectromyographic Investigation of Tensed and Relaxed Subjects. Journal
of Nervous and Mental Disease, 115, 512.
MCBRIDE, E. I). (1950) : Disability Evaluation. Archives of Physical Medicine, 31, 35.
McINToSH, R., BADGLEY, C. E., GHORMLEY, R. K., GUDAKUN5T, D. W., Iv�’, A. C., KARSNER, H. T.,
LEE, Ft. I., and \‘IETS, H. R. (1945) : Evaluation of the Results of Treatment in Infantile Paralysis. Journal
of the American Medical Association, 128, 21.
MAISON, (�. L., and BROEKER, A. G. (1941): A Simple and Compact Ergograph for Student Use. Journal
of Laboratory and Chincial Medicine, 26, 857.
MANOUVRIER, L. (1915) : Sumr he mesure des restaurations motriceset muscuhaireschezles blesses convalescents.
Comptes rendus hebdomadaires des Seances et M#{233}moires de ha Soci#{233}t#{233}de Biologic, 78, 239.
MARBLE, H. C. (1920) : Application of Curative Therapy in the Ward. Journal of Orthopaedic Suirgery,
N.S. 2, 136.
MARTIN, E. G. (1921) : Tests of Muscular Efficiency. Physiological Reviews, 1, 454.
MARTIN, E. G., and LOVETT, R. W. (1915) : A Method of Testing Muscular Strength in Infantile Paralysis.
Jouirnah of the American Medical Association, 65, 1,512.
MARTIN, E. (;., and RICH, W. H. (1918) : Muscular Strength and Muscular Symmetry in Human Beings.
American Jouirnah of Physiology, 47, 29.
MAYER, L., and GREENBERG, B. B. (1942): Measurements of the Strength of Trunk Muscles. Journal of
Bone and Joint Surgery, 24, 842.
MILcH, H. (1945) : Measurement of Muscle Strength. Journal of Bone and Joint Surgery, 27, 137.
MOLANDER, C. 0., and WEINMANN, B. (1942) : A Repetitive Resistance Test for Muscular Fatigue. Archives
of Physical Therapy, 23, 276.
MosSo, A. (1890): Les lois de la fatigume #{233}tudi#{233}esdans hes muscles de h’homme. Archives Itahiennes de
Biologic, 13, 123.
M.R.C. War Memorandum No. 7 (1942) : Aids to the Investigation of Peripheral Nerve Injuries.
MULLER, E. A. (1935) : Die gunstigste Anordnung im Sitzen betatigter Fusshebeh. Arbeitsph\’siologie, 9, 125.
NELSON, N. (1947) : Factors to be Considered in Evaluating Effect of Treatment in Anterior Pohiomvehitis.
Archives of Physical Medicine, 28, 358.
NEWMAN, L. B. (1949) : A New Device for Measuring Muscle Strength : the Myometer. Archives of Physical
Medicine, 30, 234.
NUTTER, J. A. (1919): The Standardization of Joint Records. Journal of Orthopaedic Surgery, N.S. 1, 423.
PARKER, J. S. (1929) : Recording Arthroflexometer. Journal of Bone and Joint Surgery, 11, 126.
PATRICK, J. (1946) : A Study of Supination and Pronation with Especial Reference to the Treatment of
Forearm Fractumres. Journal of Bone and Joint Surgery, 28, 737.
PROVINS, K. A. (1955): Maximum Forces Exerted About the Elbow and Shoulder Joints on Each Side
Separately and Simuhataneouisly. Journal of Applied Physiology. 7, 390.
PROVINS, K. A., and SALTER, N. (1955): Maximum Torque Exerted About the Elbow Joint. Journal of
Applied Physiology, 7, 393.
RALSTON, H. J., INMAN, V. T., STRAIT, L. A., and SHAFFRATH, M. D. (1947): Mechanics of Human Isolated
Volumntar Muscle. American Journal of Physiology, 151, 612.
THE JOURNAL OF BONE ANI) JOINT SURGERY
METHODS OF MEASUREMENT OF MUSCLE AND JOINT FUNCTION 491
RoBINsoN, \V. H. (1921) : Joint Range. Journal of Orthopaedic Surgery, N.S. 3, 41.
ROSI�N, N. G. (1922): A Simplified Method of Measuring Amplitude of Motion in Joints. Journal of Bone
and Joint Surgery, 4, 570.
RUDI), J . L. (1951) : A Dvnamometer and Exerciser. Archives of Physical Medicine, 32, 347.
RUSSELL, W. Ritchie (1952) : Pohiomyelitis, Chapter IX. London Edward Arnold & Co.
SALTER, N., and DARCUS, H. D. (1952) : The Effect of the Degree of Elbow Flexion on the Maximum Torques
I)eveloped in Pronation and Supination of the Right Hand. Journal of Anatomy, 86, 197.
SALTER, N., and DARCUS, H. D. (1953) : The Amplitude of Forearm and of Humeral Rotation. Journal of
Anatomy, 87, 407.
SCHMIER, .�. A. (1939): A Muscle Tester for Poliomyelitis Cases. Journal of Bone and Joint Surgery, 21, 475.
SCHMIER, .�. A. (1945) : Research \\‘ork on a More Precise Method of Determining Muscle Strength in
Pohiomrmvclitis Patients: a New Muscle Tester. Journal of Bone and Joint Surgery, 27, 317.
SCHWAB, R. S. (1953): Motivation in Measurements of Fatigue. In Symposium on Fatigue, p. 143. Edited
by \V. F. Fho�d amid A. T. Wehford. London: H. K. Lewis & Co. Ltd.
ScHwAII, H. S., WATKINS, A. I.., and BRAZIER, M. A. B. (1943) : Quantitation of Muscumlar Function in
Cases of Poliomyehitis and Other Motor Nerve Lesions. Archives of Neurology and Psychiatry, 50, 538.
SIL\’ER, I). (1923) : Measurement of the Range of Motion in Joints. Journal of Bone and Joint Surgery, 5, 569.
SIMoN, T., and SIMONNET, H. (1938) : Description dun ergographe. Travail humain, 6, 52.
SINELNIKOFF, E., and GRIG0R0wITScH, M. (1931) : Die Beweghichkeit der Gehenke als sekund#{228}res
geschlechtliches und konstitutionelles Merkmal. Zeitschrift f#{252}rdie gesamte Anatomic, Anteihuing 2, 15, 679.
SONNTAG, C. F. (1917): Demonstration of Ergograph. Proceedings of the Royal Society of Medicine (Section
of Balneohogv amid Climatology), 10, 69.
STARR, I. (1951) : Units for the Expression of Both Static and I)ynamic Work in Similar Terms, and thmeir
Application to \Veight-Lifting Experiments. Journal of Applied Physiology, 4, 21.
1ARBELL, L. A. (1950) : A New I)igit Myometer. Archives of Physical Medicine, 31, 159.
TAYLOR, H. L., and BROZEK, J. (1944): Evaluation of Fitness. Federation Proceedings, 3, 216.
TUTTLE, W. \V., JANNEY, C. D., and THo�lPsoN, C. W. (1950): Relation of Maximum Grip Strength) to
Grip Strength Endurance. Journal of Applied Physiology, 2, 663.
\ERNON, H. M. (1924): The Influence of Rest-Pauses and Changes of Posture on the Capacity for Muscimlar
\\‘ork. Indulstrial Fatigue Research Board Report No. 29, part B, p. 28.
W’AKIM, K. G., GERSTEN, J. \V., ELKINS, E. C., and MARTIN, G. M. (1950): Objective Recording of Muscle
Strength. .�rchives of Physical Medicine, 31, 90.
\\EDDELL, (;., and 1)ARcUS, H. D. (1947) : Some Anatomical Problems in Naval \Varfare. British) Journal
of Indumstriah Medicine, 4, 77.
WEST, C. C. 1945): Measurement of Joint Motion. Archives of Physical Medicine, 26, 414.
W’IECHEC, F. J., and KRUSEN, F. H. (1939): A New Method of Joint Measurement and a Review of the
Literatumre. .�merican Journal of Sumrgery, N.S. 43, 659.
\\ILKIE, I). R. (1950) : The Relation Between Force and \‘elocitv in Human Muscle. Journal of Ph�’siologv,
110, 249.
\\ILMER, H A., and ELKINS, E. C. (1947): An Optical Goniometer for Observing Range of Motion of Joints.
Archives of Physical Medicine, 28, 695.
\\ILS0N, (. I)., and STASCH, \V. H. (1945): Photographic Record of Joint Motion. Archives of Physical
Medicimie, 26, 361.
ZANKEL, H. T. (1951): Photogoniometry. Archives of Physical Medicine, 32, 227.
ZIN0vIEFF, A. N. (1951): Heavy Resistance Exercises: the” Oxford Technique.” British Journal of Physical
Medicine, N.S. 14, 129.
voL. 37 B, NO. 3, AUGUST 1955